STUDY OF CORRELATION OF STAGES OF ACUTE KIDNEY INJURY AND CORRELATION WITH SHORT- AND LONG-TERM OUTCOMES

Acute kidney injury (AKI) is defined as “impairment of kidney filtration and excretory function over days to weeks, resulting in the retention of nitrogenous waste products normally excreted by the kidney.” Acute kidney injury (AKI) is a well-known complication that affects critically ill patients in intensive care unit and is associated with increased mortality, morbidity, and length of stay. The incidence of AKI is extremely variable between 2.5% and 92%, which necessitate earlier detection and management to decrease risk for death, prolonged hospitalization, and future development of chronic kidney disease (CKD).

The aim of the study is to assess the short- and long-term outcomes of patients admitted with different stages of acute kidney injury. This study was a prospective observational study done in Osmania General Hospital, Hyderabad, India. Study period was 20 months and 212 patients admitted with Acute Kidney Injury of age 18-80 years were included in the study. 

Exclusion criteria:

All cases with previously documented chronic kidney disease (glomerular filtration rate [GFR] <60 ml/1.73 m2, proteinuria, and abnormal renal imaging for ≥3 months)

Solid organ transplant recipients.

History of receiving renal replacement therapy before admission and readmissions to ICU

were excluded from the study.

Patients not willing for renal replacement therapy.

Patients lost to follow-up.

Patients not willing to participate in the study.

The mean age was 45.22±14.9 years. Majority of the patients were in 5th and 6th decade accounting for 45.3% (96/212). Males dominated the study population with male: female ratio of 1.3:1 (119/93). Of the multiple etiologies of AKI, in our study population the most common causes are Sepsis (30%), hypovolemia (16.5%) and obstructive uropathy (12.3%). Most common causes of sepsis in the study population are Urosepsis accounting for 43% followed by soft tissue infections and LRTI. Pregnancy related AKI accounts for 8.5% of the study population (18/212). 27% of the study population had Diabetes/Hypertension. In the study population, the majority had Intrinsic AKI accounting for 66% (140/212). 

Pre-renal AKI was seen in 21.2% (45/212) of the patients; and rest account for post-obstructive AKI (12.7%, 27/212). Community acquired AKI was the common type of AKI in the study population accounting for 71.2% (151/212); whereas 61 developed Hospital acquired AKI (28.8%, 61/212). Most of the study population had stage-III AKI accounting for 71% (150/212) followed by stage-II AKI in 20% patients (42/212) and 20 patients had stage-I AKI (20/212). 

Hemodialysis was done in 92 patients, whereas hemodialysis alone was done in 66 patients and mean hemodialysis sessions were 4.55±2.3 during hospital stay.Acute peritoneal dialysis was done in 45 patients, but acute peritoneal dialysis alone was done in 24 patients and mean peritoneal dialysis cycles during hospital stay were 42.4±18.24 cycles. Continuous renal replacement therapy was done in 7 patients during hospital stay and all 7 patients received 1 session each, but in 5 patients’ hemodialysis was initiated after 1 session of CRRT. 54 patients underwent renal biopsy accounting for 25.5% (54/212) of which acute tubular injury (16/54) followed by IRGN (6/54) and ATIN (6/54) were the most common histological diagnoses encountered.

MEAN HOSPITAL STAY:  9.61±5.65 DAYS.  At the time of discharge, 33 patients were completely recovered; 126 patients partially recovered, and 20 patients didn’t recover even partially. 33 patients expired during hospital stay. 179 patients were there at 30-day follow-up. 72 patients completely recovered, 85 patients partially recovered, and 13 patients didn’t recover. 9 patients expired in 1 month after discharged from hospital. 170 patients were on follow-up at 3 months.  Of which, 77 patients completely recovered, 88 patients were labelled as CKD according to calculated eGFR based on CKD-EPI CREATININE equation, and 9 patients expired from 30-day to 90-day follow-up. At the end of 1 year, of 212 patients, 74 patients recovered completely accounting for 35% (74/212); 74 patients progressed to CKD accounting for 35% (74/212), and 64 patients expired accounting for 30% (64/212). Of 74 patients who progressed to CKD, 17 patients were in CKD-3A (23%, 17/74), 19 patients in CKD-3B (25.7%, 19/74), 22 patients in CKD-4 (29.7%,22/74) and 16 patients in CKD-5 (21.6,16/74) of which 6 patients were in ESRD.

     I.          URINE NGAL IN AKI:

Urine NGAL was done in 50 AKI patients in our study taking 50 normal population as controls as a part of research. Mean age and gender ratio were comparable. Of 50 patients, 18/50 were at risk of AKI (h/o contrast exposure, h/o nephrotoxic drugs); 32 patients had established AKI (4/32 had PRE-RENAL AKI, 28/32 had INTRINSIC AKI). 6/32 patients were in stage-1 AKI, 6/32 were in stage-2 AKI, and 20/32 were in stage-3 AKI. Urine NGAL was significant in assessing PRE-RENAL vs INTRINSIC AKI (119.5 ng/ml vs 1440.5 ng/ml, p=0.04). urine NGAL was significant in patients with Established AKI vs AT-RISK patients (875.5 ng/ml vs 50 ng/ml, p=0.01). Urine NGAL was high in stage-3 AKI vs stage-2 and stage-1 AKI. (1856.5±1534.9 ng/ml vs 679.2±1161.7 ng/ml vs 143.3±161.6 ng/ml). Urine NGAL was significantly high in patients treated with RRT vs conservative treatment (1750ng/ml vs 176 ng/ml, p<0.01). Urine NGAL was significantly high in patients who expired vs discharged (2100 ng/ml vs 850 ng/ml). Urine NGAL significantly correlated with serum creatinine (pearson correlation-0.540, p=0.005) and BUN/cr ratio (pearson correlation- -0.496, p=0.04). To conclude, uNGAL values were statistically significant in differentiating Renal vs Pre-renal AKI, statistically significant in predicting need of Renal Replacement Therapy, significant correlations between uNGAL and serum creatinine values, uNGAL and BUN/cr ratios, uNGAL values were statistically significant in predicting in-hospital mortality. Further large sample studies are needed to use urine NGAL in guiding management of AKI.

Our study included a total of 212 patients of which 56.1% were males and 43.9% were females, with mean age of study population being 45.22±14.91 years. SEPSIS is the most common cause of AKI in the present study as compared to other Indian and Western studies. 71% of our study population had CAAKI, whereas 29% had HAAKI. 54 patients were evaluated with renal biopsy whenever indicated, of which most common histological diagnosis was ATIN. 33 patients expired during hospital stay, 159 patients recovered either completely or partially and 20 patients didn’t recover at the time of discharge (SHORT-TERM OUTCOME).

During the hospital stay, 42% of patients required renal replacement therapy, of which 66 patients required intermittent hemodialysis (iHD) alone, 24 patients were treated with peritoneal dialysis (PD) alone, 21 patients were treated with both iHD and PD, 7 patients were treated with CRRT. Factors that had significant impact on short-term outcome were mean age of presentation, serum albumin, community acquired AKI, Intrinsic AKI, need of renal replacement therapy, severity of AKI, need of mechanical ventilation and ionotropic support during hospital stay. 74 patients completely recovered at the end of 1-year, whereas 74 patients progressed to CKD of which 8 patients were RRT dependent, and 64 patients expired at the end of 1-year follow-up. (LONG-TERM FOLLOW-UP).

Factors that had significant impact on long-term outcome were mean age at presentation, presence of comorbidities like Diabetes (p=0.001), Hypertension (p=0.03), CAD/CVA (p=0.04), DCLD (p=0.001); mean serum albumin (p=0.001); need of ionotropic support (p=0.001), mechanical ventilation during hospital stay (p=0.001); community acquired AKI (p=0.001); need of RRT during hospital stay (p=0.001); severity of AKI (p=0.001).

Urine NGAL were statistically significant in differentiating Renal vs Pre-renal AKI, statistically significant in predicting need of Renal Replacement Therapy, significant correlations between uNGAL and serum creatinine values, uNGAL and BUN/cr ratios, uNGAL values were statistically significant in predicting in-hospital mortality. Further large sample studies are needed to use urine NGAL in guiding management of AKI.

To conclude, Prompt management of AKI is required as it is associated with significant in-hospital mortality and long-term morbidity and mortality.

Implementation of AKI care bundles

Optimization of intravascular volume status and hemodynamics

Avoidance or prompt withdrawal of nephrotoxic drugs

Prevention and prompt maintenance of hyperglycemia

May help in preventing progression of AKI and reducing adverse outcomes.

Community programs should be conducted to avoid the use of nephrotoxins, NSAIDS, plant toxins and in-hospital rigorous monitoring of at-risk patients should be done to prevent development of acute kidney injury.

Regional as well as country wide registries are required to identify the common and preventable causes of AKI and necessary steps to be taken for prevention.